[PATCH nft 1/8] tests: adjust output to silence warnings

[Florian Westphal <fw@strlen.de>]

silence following (correct but harmless) warnings:
bridge/icmpX.t: WARNING: line: 6: 'src/nft add rule --debug=netlink bridge test-bridge input icmp type echo-request': 'icmp type echo-request' mismatches 'ether type ip icmp type echo-request'
bridge/icmpX.t: WARNING: line: 8: 'src/nft add rule --debug=netlink bridge test-bridge input icmpv6 type echo-request': 'icmpv6 type echo-request' mismatches 'ether type ip6 icmpv6 type echo-request'
inet/icmpX.t: WARNING: line: 6: 'src/nft add rule --debug=netlink inet test-inet input icmp type echo-request': 'icmp type echo-request' mismatches 'meta nfproto ipv4 icmp type echo-request'
inet/icmpX.t: WARNING: line: 8: 'src/nft add rule --debug=netlink inet test-inet input icmpv6 type echo-request': 'icmpv6 type echo-request' mismatches 'meta nfproto ipv6 icmpv6 type echo-request'

in all of these cases, we *could* remove the dependency, it can
be correctly re-built using icmp/icmpv6.

However, at this time, nft dependency removal does not have the needed
information to do this correctly.  In order to remove the ll dependency
(ether type, meta nfproto) we would need to know if the layer 4 protocol
is icmp (implies ipv4) or icmpv6 (implies ipv6).

Access to the next expression (meta l4proto) is NOT enough, for example:
ether type ip meta l4proto tcp

does not imply ip, we would need access to the rhs (the layer 4
protocol number) to know if the layer 2 check is (was) implied by another
statement later on.

To do that we would need two passes over a rule, or we would need to
track dependencies per-base.

So for now just accept that we don't handle it.

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 tests/py/bridge/icmpX.t | 4 ++--
 tests/py/inet/icmpX.t   | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/py/bridge/icmpX.t b/tests/py/bridge/icmpX.t
index 4d7b9b0637aa..58e366c00712 100644
--- a/tests/py/bridge/icmpX.t
+++ b/tests/py/bridge/icmpX.t
@@ -3,6 +3,6 @@
 *bridge;test-bridge;input
 
 ip protocol icmp icmp type echo-request;ok;icmp type echo-request
-icmp type echo-request;ok
+icmp type echo-request;ok;ether type ip icmp type echo-request
 ip6 nexthdr icmpv6 icmpv6 type echo-request;ok;ip6 nexthdr 58 icmpv6 type echo-request
-icmpv6 type echo-request;ok
+icmpv6 type echo-request;ok;ether type ip6 icmpv6 type echo-request
diff --git a/tests/py/inet/icmpX.t b/tests/py/inet/icmpX.t
index 43ac0909195f..91f7b9e1c472 100644
--- a/tests/py/inet/icmpX.t
+++ b/tests/py/inet/icmpX.t
@@ -3,8 +3,8 @@
 *inet;test-inet;input
 
 ip protocol icmp icmp type echo-request;ok;icmp type echo-request
-icmp type echo-request;ok
+icmp type echo-request;ok;meta nfproto ipv4 icmp type echo-request
 ip6 nexthdr icmpv6 icmpv6 type echo-request;ok;ip6 nexthdr 58 icmpv6 type echo-request
-icmpv6 type echo-request;ok
+icmpv6 type echo-request;ok;meta nfproto ipv6 icmpv6 type echo-request
 # must not remove 'ip protocol' dependency, this explicitly matches icmpv6-in-ipv4.
 ip protocol ipv6-icmp meta l4proto ipv6-icmp icmpv6 type 1;ok;ip protocol 58 meta l4proto 58 icmpv6 type destination-unreachable
-- 
2.13.6